The general goal of the proposed research is to understand how humans process dynamic changes in binaural cues. Specifically, the research is aimed toward an understanding of how the detection and discrimination of dynamically changing binaural cues is affected by the spectral and temporal characteristics of the stimuli. We propose (Aim 1) to conduct a series of behavioral experiments using stimulus conditions chosen to elucidate how and under what conditions leading and/or trailing binaural information affects the detection and discrimination of short-lived and/or dynamically changing binaural cues. The set of experiments is designed to test the notion that the detectability and discriminability of binaural cues depends upon spectral and/or temporal contrasts available in the (across-frequency) cross-correlation function. We believe that the new empirical data, combined with our new theoretical insights, will allow us to reconcile the results of apparently contradictory findings in the literature. We also propose (Aim 2) to use a cross-correlation based model of binaural hearing that incorporates peripheral auditory processing in order to provide a quantitative, mechanistic account of phenomena typically described via temporal windows and the new data obtained under Aim 1 concerning how temporal and spectral contrasts influence binaural detectability. The potential health benefits of this program of research are a better understanding of how thc ear and brain process information and the potential for "better" diagnostic procedures that may, eventually, have clinical significance.